Fractional-Order PIλDµ Cost Function-Based Model Predictive Current Control for PMSM
摘要
To address the issues of reduced current control accuracy, increased ripple, and deteriorated system stability caused by parameter mismatch in finite-state model predictive current control (FCS-MPCC) for permanent magnet synchronous motors (PMSM), this study proposes a fractional-order PIλDµ (FOPID) type cost function. By introducing fractional-order calculus operators to enhance the system’s robustness against parameter disturbances, we integrate the optimization characteristics of model predictive control with the dynamic adjustment capability of FOPID. A fractional-order calculus component for current tracking error is constructed in the cost function to effectively suppress steady-state errors and harmonic oscillations caused by parameter mismatch. MATLAB/Simulink simulations demonstrate that compared to traditional FCS-MPCC and PID-based cost function methods, the proposed FOPID-FCS-MPCC strategy can effectively extend the system bandwidth, significantly reduce current ripple and harmonic distortion (THD) under both matched and mismatched parameter conditions, and improve overall system stability. Furthermore, across a wide speed range encompassing both low-speed and high-speed operations, the FOPID-FCS-MPCC exhibits excellent anti-interference capability, load-bearing capacity, and dynamic regulation performance.